I definitely agree with your point that sampling populations throughout the US that attend mental health clinics that focus on trauma-related disorders would be a great future direction. It appears that their sample is quite limited and there is also some bias in their sample, specifically Study 2. I believe they sent out surveys to UCLA undergraduates that are taking a particular course for credit. No information was provided on what course that was and whether that course is typically predominantly one particular gender or race/ethnicity. Maybe extending that survey to other courses that are more heterogeneous in nature would be reasonable way to get a more representative sample?